Why is Data Integrity So Important?
Written by Richard Delhaye, Senior Science Consultant
I was recently asked a question about my thoughts on when it is appropriate to begin meeting the expectations of Good Documentation Practices (GDP) and data integrity in a development lab. My answer? Day one. Of course, that response got a good laugh from the person asking me this. The lab I was working with had no intention of ever doing regulated sample analysis or falling under the cGMP manufacturing guidelines.
After spending time with them and learning about their business model, I came to understand how different and difficult it is for some scientists, lab managers, and directors from the 90s (and even the early 2000s) to fully realize how significantly this landscape has changed.
Previously, a lab could execute development work on an assay method, characterize the assay, and have it validated in a GMP/GLP (GxP) laboratory. The GxP laboratory would be responsible for following GDP requirements—and everyone was happy. But those days are gone.
Data Integrity and the FDA
Data integrity principles (ALCOA/+/++) are now the backbone of GDP. In fact, in terms of the FDA, data quality and management/life cycle are now front and center.
In 2018, the finalized guidance from the FDA on data integrity was released, which clarified the role of data integrity in manufacturing (cGMP) for drugs in relation to 21 CFR parts 210, 211, and 212. In footnote 14 in the final guidance, the FDA indicates that pursuant to 21 USC § 374(a), the agency may inspect “records not intended to satisfy a CGMP requirement, but which nonetheless contain CGMP information.”
This statement allows the FDA to request any documentation that can be used to reconstruct an activity. This means that they could fully inspect a development lab’s documentation, look for thorough explanations of who, how, and why an assay was developed, and determine if initial stability was performed in development. Ultimately, the FDA will expect the documentation to meet the ALOCOA/+/++ expectations.
The Effects of Insufficient Documentation
In one of my previous positions, I was asked to bring up an old cell-based bioassay. After a few weeks of searching 10-year-old research notebooks from the archive (followed by a week-long search through a liquid nitrogen freezer), I found the cell line. Once I had confirmed the method for bringing up the cells, expanding them, and creating a new cell library, I was ready to start. And after six months of work, I was able to reproduce the original development assay. We then started to use it for research and development—in fact, the assay was working so well that we performed an abbreviated qualification.
When management found out we saved >250K in the research and development side they wanted us to finalize development and move it over to be used for commercial release. But there was a problem. After going through the notebooks, I knew there were a lot of documentation issues and a general lack of data. I also knew that updating the assay for use in a GMP environment would be difficult. I had a lot of discussions with quality about what would be required to fully validate the assay and learned that the lab would need to be brought up to current ICH guidelines for cell culture laboratories. Additionally, the assay and all critical components would need to be described and shown to be under control for cGMP regulations. While these updates occurred, I started the work on the components and reagents.
As I was updating the method and documenting the development, I found that the cell line used had just appearedin the lab—no one could explain how it got to us, or how the cells had been transfected. After some searching, I was able to find the scientist who brought them into the lab; but unfortunately, they had retired and passed away a few years after leaving. So, the single most important component of the assay had no documented history or critical manufacturing data. As a result, I was left with an assay that we knew would work great and would pass a GMP validation, but my critical reagent was unusable because of a lack of good documentation.
At the beginning of development, no one involved believed it was ever going to move over to the manufacturing side and be part of the commercial release of the final product. So, they did not treat the collection, storage of data, and documentation of the invention/creation of reagents to an appropriate level.
I ended up spending a little less than a year sourcing the cell line and getting the appropriate rights to use it for commercial use. All this time would have been saved if the scientist had followed basic good documentation practices. And, if the early development had been documented properly, the company would have saved a million dollars in assay development costs—at a minimum.
The Depth of Data Integrity I think most organizations believe they understand what GDP is because they have been in the life sciences for so long. But the definition of data has changed considerably over the last decade and is now a highly encompassing concept. Below is only a small sampling of what must be considered when looking at GDP and data integrity in laboratories:
The FDA has indicated that, in certain cases, they are willing to visit development laboratories if a critical reagent for a cGMP assay is manufactured there—is your lab ready?
How Can We Help?
Process Alliance is a leader in the GxP space for providing high quality results for our clients that are expanding or have been asked to update their internal processes by their customers. To learn more about how our experts can help improve your data integrity strategy, email us at email@example.com, or visit our website.